The major phenotypic aspects of Down Syndrome include mental retardation, leukemia, heart and gut defects, immune deficiencies, and Alzheimer's Disease. An unknown number of genes on human chromosome 21 are responsible for Down Syndrome when they are present in three copies. Fewer than 2 dozen of the probably 500-1000 genes located on chromosome 21 have been cloned and characterized. As a result, we do not have a sufficient number or variety of genes from which to choose the most credible and compelling candidate genes for the spectrum of Down Syndrome features. We have used the physical and phenotypic map of chromosome 21 to choose regions and clones (YACs, cosmids and lambda) that are likely to contain relevant genes. We propose to: i) isolate genes coding for transcription factors and cell surface markers, using DNA sequence motifs and tissue specific antibodies, ii) isolate genes from the region spanning approximately 3 mbp surrounding D21S55, because this region is likely involved in many of the most medically critical features, iii) search for new genes in 21q21, because this region is associated with Alzheimers Disease. To characterize these genes, we propose to: i) analyze sequence information through GenBank and Grail, ii) determine tissue specificity and developmental timing of expression in human and mouse embryonic, fetal and adult tissue, and iii) use a mouse partial trisomy 16 model system to determine dosage effects and tissue localization. Based on this information, we will choose genes judged most likely to be relevant to Down Syndrome, and begin protein characterization. The ultimate aim of this proposal is to increase our understanding at the gene level of developmental events, as they apply to Down Syndrome. In the process, we will generate a well characterized collection of genes from which multigene constructs for the creation of transgenic can eventually be developed.